Postdoctoral Fellow - AI for Medical Image Analysis and Generative Models

The Quantitative Intelligent Imaging Lab (QUINlab) at Boston Children's Hospital and Harvard Medical School is seeking a highly motivated Research Fellow or Research Scientist to join an NIH R01-funded research program developing advanced AI methods for pediatric medical imaging. The position will focus on machine learning and computer vision for clinically important imaging problems, including generative AI, medical image analysis, and learning from limited, heterogeneous, or imbalanced datasets. The candidate will work closely with radiologists, clinicians, computer scientists, and clinical researchers to develop and evaluate clinically meaningful AI tools. Key responsibilities: Develop and evaluate deep learning methods for medical imaging and computer vision. Investigate generative AI approaches for image synthesis and data augmentation. Develop algorithms for vision-language modeling, image classification, object detection and localization, segmentation, and related tasks. Explore annotation-efficient, weakly supervised, self-supervised, or foundation-model approaches where appropriate. Evaluate model robustness and generalizability across datasets and institutions and against expert annotations. Prepare analyses, figures, manuscripts, conference abstracts, and grant progress reports. Maintain reproducible code and documentation suitable for internal use and open-source release where appropriate. Required qualifications: PhD or equivalent advanced degree in electrical engineering, computer science, biomedical engineering, data science, computational imaging, or a related field. Strong programming skills in Python. Experience with machine learning or deep learning frameworks such as PyTorch, TensorFlow, MONAI, or similar tools. Strong background in computer vision, medical image analysis, or generative AI. Ability to work independently and collaborate closely with clinicians and technical team members. Excellent organizational, scientific writing, and communication skills. High-quality publications in top journals. Preferred experience: Generative or diffusion models, object detection, segmentation, image synthesis, weak/self-supervised learning, or foundation models. Experience with medical imaging datasets, multi-institutional studies, or clinically oriented AI research. Research environment and how to apply: QUINlab's mission is to transform diagnosis and treatment through intelligent imaging and analysis. The lab is based at Boston Children's Hospital and affiliated with Harvard Medical School, within a highly collaborative biomedical research ecosystem that includes Harvard, MIT, leading academic medical centers, and a strong biotechnology and AI community. Interested candidates should send a CV, a brief statement of research interests, career goals and examples of relevant code (GitHub repos), one or two top publications and contact information for two references to sila.kurugol@childrens.harvard.edu. Location: Boston, MA. Affiliations: Boston Children's Hospital + Harvard Medical School. Start: Flexible / as soon as possible.